Automatic classification of leukocytes using deep neural network

Yu, Wei; Jing, Chang; Yang, Cheng; Zhang, Limin; Shen, Han; Xia, Yongquan; Sha, Jian

doi:10.1109/asicon.2017.8252657

Cited by 80 publications

(34 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To evaluate the capability of WBC-Profiler on leukocytes classification, we have compared it with one of the state-of-the-art techniques in the field of leukocytes classification (we refer it to as DeepVote, which combines the classification results from different deep neural networks to vote for the final decision) [23] and with one of the most successful deep learning techniques in the field of object detection and recognition, i.e., Faster R-CNN [28] with both vgg16 and res101 as the network architectures. During evaluation, half-half cross-validation was employed with 10 iterations, where, at each iteration, we randomly selected 50% of the data per cell type for training, and used the rest for testing, and the performance in terms of average F1-measure and confusion matrix was illustrated in Figure 3, which demonstrates the effectiveness of WBC-Profiler for leukocytes classification.…”

Section: Resultsmentioning

confidence: 99%

“…Motivated by recent neuroscience findings [9, 10], unsupervised learning and deep learning techniques [11–13] have gained momentum during the past decade for object representation and recognition (e.g., face representation and recognition) [14–17]. And their applications in various biomedical tasks [18–22] have demonstrated success with the potential to provide a new avenue to data-intensive clinical studies, among which, the leukocytes classification accuracy has been significantly improved due to the employment of deep neural networks (DNNs) [23]. However, systems of such kind typically only provide the end-to-end (i.e., from data to classification) solution, which leaves the characterization of white blood cells inaccessible, and thus impede the construction of leukocytes profile for many potential needs, including profile interpretation, profile optimization, profile differentiation among cell types as well as profile association with other meaningful endpoints.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

WBC-Profiler: an Unsupervised Feature Learning System for Leukocytes Characterization and Classification

Mao

Xia

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

The characterization and classification of white blood cells (WBC) is critical for the diagnosis of anemia, leukemia and many other hematologic diseases. We developed WBC-Profiler, an unsupervised feature learning system for quantitative analysis of leukocytes. We demonstrate that WBC-Profiler enables automatic extraction of complex signatures from microscopic images without human-intervention and thereafter effective construction of leukocytes profiles, which decouples large scale complex leukocytes characterization from limitations in both human-based feature engineering/optimization and the end-to-end solutions provided by modern deep neural networks, and therefore has the potential to provide new opportunities towards meaningful studies/applications with scientific and/or clinical impact 2 indication, diagnosis and evaluation of many conditions, including infection, inflammation and anemia. Among vari-3 ous hematology tests, the differential count of while blood cells (WBC) provides an important tool in diagnosing and 4 monitoring infection and leukemic disorders, and the ratio of various kinds of leukocytes are commonly used as impor-5 tant markers. For example, the neutrophil to lymphocyte ratio (NLR) is used as a marker of subclinical inflammation, 6 and recent studies suggest that increased NLR is independent predictor of mortality in patients undergoing angiogra-7 phy or cardiac revascularization [1], meanwhile it is also associated with poor prognosis of various cancers [2], such 8 as esophageal cancer [3] or advanced pancreatic cancer [4].9Accurate characterization, detection and classification of while blood cells into several categories, including Mono-10 cytes, Lymphocytes, Basophils, Eosinophils, Atypical lymphocytes and Neutrophilic granulocytes, is critical for the 11 ratio (proportion) assessment of leukocytes in blood cell slides. However, such a high-demanding task in clinical lab-12 oratory heavily relies on the manual annotation by pathologies, which is not only labor-intensive but also challenging 13 given the shortage of experienced medical experts in many clinical laboratories. To overcome these obstacles, many 14 efforts have been made towards the automatic blood cell classification. Among which, early development of com-15 mercial systems in 1970s failed to revolutionize the field due to the high price and low accuracy [5] of the products, 16 and Leuko, another commercial leukemia diagnosis system, was later on developed to reveal an improved accuracy 17 on cell classification based on naive Bayes classifiers. Meanwhile, research efforts on cell classification have also 18 1 been evolving from fuzzy logic techniques [6], support vector machines (SVMs) [7] to cellular neural networks [8]. 19 However, these works were mostly developed with a limited amount of data, and/or focused on images taken from 20 specified instruments, which leave their generalization capability insufficiently justified. 21Motivated by recent neuroscience findings [9, 10], unsupervised learning and deep le...

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

WBC-Profiler: an Unsupervised Feature Learning System for Leukocytes Characterization and Classification

Mao

Xia

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…This section covers some of the research conducted in the field of diseases, especially blood disease detection and diagnosis. Some of these studies pertain to traditional methods [8,9], which consist of several steps such as pre-processing, segmentation, feature extraction, and classification, whereas other methods pertain to deep-learning-based methods [6,[10][11][12], which use deep neural networks for end-to-end learning tasks.…”

Section: Related Studiesmentioning

confidence: 99%

Deep Transfer Learning in Diagnosing Leukemia in Blood Cells

2020

View full text Add to dashboard Cite

Leukemia is a fatal disease that threatens the lives of many patients. Early detection can effectively improve its rate of remission. This paper proposes two automated classification models based on blood microscopic images to detect leukemia by employing transfer learning, rather than traditional approaches that have several disadvantages. In the first model, blood microscopic images are pre-processed; then, features are extracted by a pre-trained deep convolutional neural network named AlexNet, which makes classifications according to numerous well-known classifiers. In the second model, after pre-processing the images, AlexNet is fine-tuned for both feature extraction and classification. Experiments were conducted on a dataset consisting of 2820 images confirming that the second model performs better than the first because of 100% classification accuracy.Computers 2020, 9, 29 2 of 12 and classification. However, the success of each step depends on the success of the preceding step. For example, the success of classification depends on the success of the preceding feature extraction, which itself depends on the success of the preceding segmentation. Hence, high classification accuracy requires the success of all steps, each of which is non-trivial and problem-dependent [1,7].Recently, deep learning achieved many breakthroughs in different fields such as computer vision, natural language processing, and object recognition. Deep neural networks, such as convolutional neural networks (CNNs), can be used effectively to build computer-aided diagnostic systems. However, the design and training of deep neural networks are non-trivial and time-consuming tasks. Hence, instead of building a deep neural network from scratch, we use the concept of transfer learning, in which a deep network that achieved success in solving a certain problem is tuned to solve another problem.This paper proposes two classification models that are based on transfer learning and can distinguish between healthy and unhealthy blood smear images with high accuracy. These models employ AlexNet, which is a deep CNN that achieved huge success in the image classification challenge, ImageNet 2012.The remaining sections of this paper are arranged as follows: Section 2 covers some of the traditional and end-to-end-based methods described in the literature. Section 3 presents our two proposed classification models. Section 4 discusses their implementation in our experiments. Section 5 discusses the results obtained. Finally, Section 6 concludes the paper.

show abstract

“…The model proposed by Yu et al [10] is a combination of state-of-the-art convolution neural networks including ResNet50, InceptionV3, VGG16, VGG19 and Xception for automatic cell recognition system using convolutional neural networks. The obtained result of the proposed model is compared to traditional machine learning algorithms such as K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Logistic Regression (LR), and Decision Tree (DT).…”

Section: Related Studiesmentioning

confidence: 99%

A Hybrid Deep Learning Architecture for Leukemic B-lymphoblast Classification

Kassani

Wesolowski

et al. 2019

2019 International Conference on Information and Communication Technology Convergence (ICTC)

View full text Add to dashboard Cite

Automatic detection of leukemic B-lymphoblast cancer in microscopic images is very challenging due to the complicated nature of histopathological structures. To tackle this issue, an automatic and robust diagnostic system is required for early detection and treatment. In this paper, an automated deep learning-based method is proposed to distinguish between immature leukemic blasts and normal cells. The proposed deep learning based hybrid method, which is enriched by different data augmentation techniques, is able to extract high-level features from input images. Results demonstrate that the proposed model yields better prediction than individual models for Leukemic Blymphoblast classification with 96.17% overall accuracy, 95.17% sensitivity and 98.58% specificity. Fusing the features extracted from intermediate layers, our approach has the potential to improve the overall classification performance.

show abstract

Automatic classification of leukocytes using deep neural network

Cited by 80 publications

References 6 publications

WBC-Profiler: an Unsupervised Feature Learning System for Leukocytes Characterization and Classification

WBC-Profiler: an Unsupervised Feature Learning System for Leukocytes Characterization and Classification

Deep Transfer Learning in Diagnosing Leukemia in Blood Cells

A Hybrid Deep Learning Architecture for Leukemic B-lymphoblast Classification

Contact Info

Product

Resources

About